Abstract Significance: Infantile Pompe Disease (IPD) is an autosomal recessive glycogen storage disorder caused by a deficiency of acid alpha-glucosidase (GAA), leading to accumulation of glycogen in lysosomes, primarily in skeletal, cardiac and smooth muscles. Untreated IPD infants die within the first 2 years of life. Enzyme replacement therapy (ERT) with recombinant human GAA (rhGAA) improves survival and quality of life, but the development of sustained anti-therapeutic antibodies (ATA) potentially reduces ERT efficacy. All IPD patients who have no GAA protein expression (circulating cross-reactive immunologic material [CRIM]-negative) are predicted to mount high titer ATA, and are treated with an intensive Immune Tolerance Induction (ITI) regimen. While some CRIM-positive IPD patients become tolerized to ERT over time, over one-third of CRIM-positive patients still develop treatment-limiting ATA. There is currently no effective method for determining which CRIM-positive patients are at risk of developing treatment-limiting ATA, and would therefore benefit from treatment with an ITI regimen. This project will validate a recently developed algorithm, GAA-iTEM, for predicting the risk of ATA development for individual IPD patients, in a retrospective study of clinical data and in vitro studies of GAA T cell epitopes. Hypothesis: CRIM-positive IPD patients are tolerized to their own residual native GAA (nGAA) but may develop ATA to ERT with rhGAA. Using well-established computational tools to identify T cell epitopes in rhGAA which are predicted to stimulate CD4 T cells (and thus promote a T dependent B cell response to rhGAA), patients may be classified as high or low risk for the development of ATA. T cell epitopes that are cross-conserved between nGAA and rhGAA (and other endogenous epitopes) are unlikely to induce a T cell response. GAA-iTEM currently predicts ATA risk, using the patient?s HLA type and nGAA sequence. We propose to improve GAA-iTEM using newly accessible clinical data and samples. In Specific Aim 1, 40 CRIM+ IPD patients followed at Duke with defined ATA responses will be evaluated using GAA-iTEM to generate a patient-specific GAA-Individualized T cell Epitope Measure (GAA-iTEM) score. Clinical and genetic factors will be used to refine the predictive tool, which will be made available through a prototype graphical user interface to Duke clinicians, who will assess patient-specific risk and deposit clinical data for future calibration of the prediction algorithm. In Specific Aim 2, GAA CD4 T cell effector and regulatory epitopes identified in silico will be evaluated for HLA binding, cytokine stimulation, and generation of Tregs. Overall Impact: GAA-iTEM will provide researchers with an individualized assessment of a patient?s risk for developing treatment-limiting ATA based on their HLA haplotype and nGAA mutations, improving decision-making regarding ITI treatment. Once validated, the new GAA-iTEM would be used to assess ATA risk in a prospective study, and to develop ATA-risk assessment tools for related genetic disorders.